Rationale and Research Questions

The U.S. energy system is undergoing a major structural shift as renewable energy, particularly wind and solar, continues to expand across the country. According to the U.S. Energy Information Administration (EIA), renewable energy sources accounted for about 21 percent of total U.S. utility-scale electricity generation in 2023, and that share continues to grow as wind, solar, and other renewables expand their role in the energy mix (EIA, 2024). Although natural gas and petroleum remain the dominant energy sources, the rapid expansion of wind and solar has become one of the most important trends shaping the nation’s electricity mix. This study highlights the regions driving the transition and shows how quickly renewable technologies are reshaping the U.S. electricity system by examining where and when new capacity has been added, using national wind and solar capacity data from 2018–2023 to analyze these trends.

Understanding how renewable capacity is changing spatially and temporally is essential for assessing the pace of the U.S. energy transition. Facility-level datasets from the EIA Form 860 and the EPA’s eGRID provide detailed information on generator technology, capacity (MW), location, and in-service dates, which can be used for analyzing both annual growth patterns and geographic distribution of renewable development.

Research questions

  1. Which U.S. states have experienced the fastest growth in renewable energy capacity from 2018 to 2023, and how does this growth vary by technology?

  2. Is renewable energy capacity growth spatially clustered across the United States, and do different technologies exhibit distinct geographic patterns of expansion?

  3. Does early adoption predict faster renewable expansion?

Dataset Information

Data Source and Collection

This analysis uses data from the Emissions & Generation Resource Integrated Database (eGRID) published by the U.S. Environmental Protection Agency (EPA) for the years 2018–2023. The datasets were obtained from the EPA eGRID archive as a series of Microsoft Excel files corresponding to each reporting year. eGRID compiles electricity generation and capacity data reported annually by U.S. power plants to the U.S. Energy Information Administration (EIA), primarily through EIA Forms 860 and 923, into a standardized national database. To support spatial analysis, this study also incorporates the 2018 Cartographic Boundary File for U.S. counties and county-equivalent units produced by the U.S. Census Bureau. This polygon shapefile provides generalized county boundaries for the entire United States and its territories and is intended for small-scale thematic mapping. The dataset reflects administrative boundaries as of January 1, 2018, is referenced to NAD83, and was used as a spatial framework for aggregating and visualizing plant-level eGRID data.

Data Content and Structure

Each eGRID Excel file contains multiple worksheets, including unit-level, generator-level, plant-level, and state-level tables. This study uses the plant-level (PLNT) table, which provides plant-level information of electricity generation and capacity across the United States, including spatial information, fuel-specific energy contributions, and capacity information. Because each plant record also includes spatial identifiers such as county code and latitude/longitude, this dataset is perfect for linking generation and capacity measures to spatial patterns across regions.

Key variables extracted for analysis include reporting year, plant location, fuel or technology classification, installed capacity (MW), and net generation (MWh). Fuel and technology classifications were used to identify renewable energy facilities, with a focus on wind, solar, and hydroelectric power. Installed capacity represents the maximum rated output of a plant, while net generation reflects actual electricity produced during each reporting year, enabling both spatial and temporal analysis of renewable energy patterns.

Data Wrangling and Preparation

Due to the large size of the raw eGRID files, initial data preparation was performed in Excel prior to analysis in R. Plant-level tables for each year (2018–2023) were converted to CSV format, and only variables required for the analytical objectives were retained due to limited storage on GitHub. To support different analytical goals, two processed versions of the dataset were created and stored in the repository (Data/Raw):

  • Time-series dataset: retained columns related to reporting year, fuel or technology category, capacity, generation, and other temporal attributes.

  • Spatial dataset: retained plant location and geographic identifiers, including state, latitude, longitude, plant identifiers, and renewable classification.

Unused variables were removed to reduce file size and accommodate repository storage constraints. Missing values were assessed, particularly in coordinate and fuel-type fields, and records lacking essential information were excluded from aggregation. Data were subsequently summarized by fuel type and location to support both temporal trend analysis and spatial visualization.

Exploratory Analysis

Spatial Exploratory Analysis

1. Explore all US counties.

Figure 1. Interactive Map. County boundaries of the United States displayed from a county-level shapefile. The map allows zooming and panning to explore spatial variation in county size and distribution. Data source: U.S. Census Bureau TIGER/Line Shapefiles

2. 2018 Electricity Plants Locations

Figure 2. Locations of electricity generating plants in the contiguous United States in 2018. Power plants are shown as blue points. Data source: EPA eGRID (2018).

Figure 2. Locations of electricity generating plants in the contiguous United States in 2018. Power plants are shown as blue points. Data source: EPA eGRID (2018).

Figure 3. Locations of electricity generating plants in Alaska, United States in 2018. Power plants are shown as blue points. Data source: EPA eGRID (2018).

Figure 3. Locations of electricity generating plants in Alaska, United States in 2018. Power plants are shown as blue points. Data source: EPA eGRID (2018).

Figure 4. Locations of electricity generating plants in Hawaii, United States in 2018. Power plants are shown as blue points. Data source: EPA eGRID (2018).

Figure 4. Locations of electricity generating plants in Hawaii, United States in 2018. Power plants are shown as blue points. Data source: EPA eGRID (2018).

3. Catogorized Electricity Plant Type

Figure 5. Locations of electricity generating plants in the contiguous United States in 2018, categorized by primary fuel type. Data source: EPA eGRID (2018).

Figure 5. Locations of electricity generating plants in the contiguous United States in 2018, categorized by primary fuel type. Data source: EPA eGRID (2018).

Figure 6. Locations of electricity generating plants in Alaksa, USA in 2018, categorized by primary fuel type. Data source: EPA eGRID (2018).

Figure 6. Locations of electricity generating plants in Alaksa, USA in 2018, categorized by primary fuel type. Data source: EPA eGRID (2018).

Figure 7. Locations of electricity generating plants in Hawaii, USA in 2018, categorized by primary fuel type. Data source: EPA eGRID (2018).

Figure 7. Locations of electricity generating plants in Hawaii, USA in 2018, categorized by primary fuel type. Data source: EPA eGRID (2018).

4. Map the pattern of total generator per state

Figure 8. Interactive Map. Total number of electricity generators by state in the United States (2018). Generator totals are aggregated from plant-level data and visualized using a color gradient. Data sources: EPA eGRID (2018); U.S. Census Bureau TIGER/Line Shapefiles.

5. Total Installed Renewable Capacity by Fuel Type

Graph 1. Total Installed Renewable Capacity by Fuel Type from 2018 to 2023. Different fuel type are shown as different colors. Data source: EPA eGRID (2018-2023).

Graph 1. Total Installed Renewable Capacity by Fuel Type from 2018 to 2023. Different fuel type are shown as different colors. Data source: EPA eGRID (2018-2023).

Analysis

Subsection 1: How have Wind, Solar, and Hydro capacities evolved from 2018–2023?

1. Visualization

Graph 2. Renewable Energy Expansion: Capacity and Generation Growth Across Technologies (2018–2023). The dotted lines indicate the smoothed trend across years for each technology’s capacity and generation additions. Data source: EPA eGRID (2018-2023).

Graph 2. Renewable Energy Expansion: Capacity and Generation Growth Across Technologies (2018–2023). The dotted lines indicate the smoothed trend across years for each technology’s capacity and generation additions. Data source: EPA eGRID (2018-2023).

Graph 2 shows how electricity generation from wind, solar, and hydroelectric sources has changed from 2018 to 2023. During this period, solar generation increased the most, reflecting the rapid growth in installed solar capacity and improvements in system efficiency. Wind generation continued to rise as well, though at a steadier pace, building on its already large share of renewable output. In contrast, hydroelectric generation remained fairly stable over these years, which aligns with the limited expansion of hydro infrastructure and the fact that hydro output is more strongly influenced by long-term facility operations and water availability than by new capacity additions.

2. Statistical analysis

Trend in Installed Nameplate Capacity (2018–2023)
Technology Slope_MW_per_yr Std_Error t_stat p_value
HYDRO -435.0143 146.3012 -2.973415 0.0410053
SOLAR 12714.7286 1135.1832 11.200596 0.0003618
WIND 10820.2914 797.5774 13.566446 0.0001709

The results from Table 1 show that solar capacity increased the fastest, growing by approximately 12.7 gigawatts per year, while wind followed closely at about 10.8 gigawatts per year. Both of these trends are statistically significant, indicating strong and consistent national expansion for solar and wind over the study period. In contrast, hydroelectric capacity shows a small but statistically significant decline of about 435 megawatts per year, reflecting the limited opportunities for new hydropower development and the gradual reduction or aging of existing facilities.

Subsection 2: How have capacity levels changed for each technology? Do different technologies exhibit distinct geographic patterns of expansion?

1. Geographic Distribution Pattern of Power Plant in USA for Wind, Hydro, and Solar Tech

Wind Power

Figure 9. 2018 Wind Power Plants - Mainland USA. Power plants are shown as brown points. Data source: EPA eGRID (2018).

Figure 9. 2018 Wind Power Plants - Mainland USA. Power plants are shown as brown points. Data source: EPA eGRID (2018).

Figure 10. 2018 Wind Power Plants - Alaska. Power plants are shown as brown points. Data source: EPA eGRID (2018).

Figure 10. 2018 Wind Power Plants - Alaska. Power plants are shown as brown points. Data source: EPA eGRID (2018).

Figure 11. 2018 Wind Power Plants - Hawaii. Power plants are shown as brown points. Data source: EPA eGRID (2018).

Figure 11. 2018 Wind Power Plants - Hawaii. Power plants are shown as brown points. Data source: EPA eGRID (2018).

Figure 18. 2023 Wind Power Plants - Mainland USA. Power plants are shown as brown points. Data source: EPA eGRID (2023).

Figure 18. 2023 Wind Power Plants - Mainland USA. Power plants are shown as brown points. Data source: EPA eGRID (2023).

Figure 19. 2023 Wind Power Plants - Alaska. Power plants are shown as brown points. Data source: EPA eGRID (2023).

Figure 19. 2023 Wind Power Plants - Alaska. Power plants are shown as brown points. Data source: EPA eGRID (2023).

Figure 20. 2023 Wind Power Plants - Hawaii. Power plants are shown as brown points. Data source: EPA eGRID (2018).

Figure 20. 2023 Wind Power Plants - Hawaii. Power plants are shown as brown points. Data source: EPA eGRID (2018).

The contrasting distribution of wind electricity plants between the mainland United States and Alaska is driven largely by differences in wind resource quality and underlying geography. The mainland map shows dense clusters of wind facilities across the Great Plains and upper Midwest: regions characterized by flat terrain and some of the most consistent, high-quality onshore wind resources in the world. These geographic and atmospheric conditions create an expansive “wind belt” extending from Texas through the Dakotas, making large-scale wind farm development both feasible and economically attractive (U.S. Department of Agriculture, 2022). In contrast, Alaska’s harsh climate and highly variable wind regimes limit where turbines can be reliably sited. Although strong winds exist along coastal and mountainous areas, much of Alaska’s landscape faces environmental and operational constraints such as icing, extreme cold, and limited accessible flat land (U.S. Department of Energy, 2021). As a result, wind development in Alaska remains sparse and localized, while the mainland supports widespread utility-scale installations.

Hydro Power

Figure 12. 2018 Hydro Power Plants - Mainland USA. Power plants are shown as dark blue points. Data source: EPA eGRID (2018).

Figure 12. 2018 Hydro Power Plants - Mainland USA. Power plants are shown as dark blue points. Data source: EPA eGRID (2018).

Figure 13. 2018 Hydro Power Plants - Alaska. Power plants are shown as dark blue points. Data source: EPA eGRID (2018).

Figure 13. 2018 Hydro Power Plants - Alaska. Power plants are shown as dark blue points. Data source: EPA eGRID (2018).

Figure 14. 2018 Hydro Power Plants - Hawaii. Power plants are shown as dark blue points. Data source: EPA eGRID (2018).

Figure 14. 2018 Hydro Power Plants - Hawaii. Power plants are shown as dark blue points. Data source: EPA eGRID (2018).

Figure 21. 2023 Hydro Power Plants - Mainland USA. Power plants are shown as dark blue points. Data source: EPA eGRID (2023).

Figure 21. 2023 Hydro Power Plants - Mainland USA. Power plants are shown as dark blue points. Data source: EPA eGRID (2023).

Figure 22. 2023 Hydro Power Plants - Alaska. Power plants are shown as dark blue points. Data source: EPA eGRID (2023).

Figure 22. 2023 Hydro Power Plants - Alaska. Power plants are shown as dark blue points. Data source: EPA eGRID (2023).

Figure 23. 2023 Hydro Power Plants - Hawaii. Power plants are shown as dark blue points. Data source: EPA eGRID (2023).

Figure 23. 2023 Hydro Power Plants - Hawaii. Power plants are shown as dark blue points. Data source: EPA eGRID (2023).

Hydroelectric plant distribution in the mainland U.S. and Alaska is shaped primarily by geography and hydrological conditions. In the mainland, hydro facilities cluster in the Pacific Northwest, the Sierra Nevada, and parts of the Appalachians—regions with steep terrain, major river systems, and abundant water flow that make large-scale hydropower feasible (U.S. Department of Energy, 1998). In contrast, Alaska’s hydro plants are concentrated along the southeastern coastline, where heavy rainfall and mountainous topography create ideal small- to medium-scale hydro sites. Much of Alaska’s interior lacks sufficient river gradients or water volume, limiting hydropower development.

Solar Power

Figure 15. 2018 Solar Power Plants - Mainland USA. Power plants are shown as orange points. Data source: EPA eGRID (2018).

Figure 15. 2018 Solar Power Plants - Mainland USA. Power plants are shown as orange points. Data source: EPA eGRID (2018).

Figure 16. 2018 Solar Power Plants - Alaska. Power plants are shown as orange points. Data source: EPA eGRID (2018).

Figure 16. 2018 Solar Power Plants - Alaska. Power plants are shown as orange points. Data source: EPA eGRID (2018).

Figure 17. 2018 Solar Power Plants - Mainland USA. Power plants are shown as orange points. Data source: EPA eGRID (2018).

Figure 17. 2018 Solar Power Plants - Mainland USA. Power plants are shown as orange points. Data source: EPA eGRID (2018).

Figure 24. 2023 Solar Power Plants - Mainland USA. Power plants are shown as dark blue points. Data source: EPA eGRID (2023).

Figure 24. 2023 Solar Power Plants - Mainland USA. Power plants are shown as dark blue points. Data source: EPA eGRID (2023).

Figure 25. 2023 Solar Power Plants - Alaska. Power plants are shown as orange points. Data source: EPA eGRID (2023).

Figure 25. 2023 Solar Power Plants - Alaska. Power plants are shown as orange points. Data source: EPA eGRID (2023).

Figure 26. 2023 Solar Power Plants - Hawaii. Power plants are shown as orange points. Data source: EPA eGRID (2023).

Figure 26. 2023 Solar Power Plants - Hawaii. Power plants are shown as orange points. Data source: EPA eGRID (2023).

The distribution of solar electricity plants in the mainland United States reflects the strong relationship between solar development and solar resource availability, climate, and land suitability. In the mainland U.S. map, solar facilities are heavily concentrated in the Southwest, including California, Arizona, Nevada, New Mexico, and Texas—regions with some of the highest solar irradiance in the country and large areas of open land well suited for utility-scale solar farms. Additional clusters appear in the Southeast and Mid-Atlantic, where supportive policies and growing energy demand have encouraged solar expansion despite slightly lower sunlight levels (National Renewable Energy Laboratory. 2018). In contrast, Alaska has no utility-scale solar plants in 2018, largely because the state receives far fewer hours of sunlight, especially in winter, and experiences long periods of low solar angle and cloudiness. These geographic and climatic constraints make Alaska unsuitable for large-scale solar development. However, there is one solar panel spotted in Alaska in 2023.

2. Statistical and Time Series Analysis

Mann–Kendall Trend Test for Installed Nameplate Capacity (2018–2023)
Technology Kendall_tau p_value signif
HYDRO -0.467 0.2597
SOLAR 1.000 0.0085 **
WIND 1.000 0.0085 **

The Mann–Kendall results presented in Table 2 reveal clear differences in long-term capacity trends across renewable technologies of Hydro, Solar, and Wind. Solar and wind both exhibit strong positive monotonic trends, with Kendall’s τ equal to 1.00 and statistically significant p-values (p = 0.0085), indicating there is a consistent increasing trend in installed capacity. In contrast, hydroelectric capacity shows a modest negative τ value (τ = –0.467), but its p-value (p = 0.2597) is not statistically significant, suggesting that there is not any statistically meaningful long-term trend. All together, solar and wind experienced clear and significant upward growth during this period, while hydroelectric capacity remained essentially stable due to geographic and infrastructural constraints.

Question 2: Subsection 3: Which states contribute most to renewable capacity growth?

1. Spatial presentation

Figure 27. Spatial Distribution of Texas Power Plants by Primary Fuel Type, 2023. Power plants are shown as colorful points. Data source: EPA eGRID (2023).

Figure 27. Spatial Distribution of Texas Power Plants by Primary Fuel Type, 2023. Power plants are shown as colorful points. Data source: EPA eGRID (2023).

Figure 28. Spatial Distribution of Washington Power Plants by Primary Fuel Type, 2023. Power plants are shown as colorful points. Data source: EPA eGRID (2023).

Figure 28. Spatial Distribution of Washington Power Plants by Primary Fuel Type, 2023. Power plants are shown as colorful points. Data source: EPA eGRID (2023).

Figure 29. Spatial Distribution of California Power Plants by Primary Fuel Type, 2023. Power plants are shown as colorful points. Data source: EPA eGRID (2023).

Figure 29. Spatial Distribution of California Power Plants by Primary Fuel Type, 2023. Power plants are shown as colorful points. Data source: EPA eGRID (2023).

2.

Top Five States in Solar Capacity Additions (2018–2023)
rank 2018 2019 2020 2021 2022 2023
1 CA CA CA CA CA CA
2 NC NC TX TX TX TX
3 AZ AZ NC NC FL FL
4 TX TX FL FL NC NC
5 FL FL AZ NV GA AZ
Top Five States in Hydroelectric Capacity Additions (2018–2023)
rank 2018 2019 2020 2021 2022 2023
1 WA WA WA WA WA WA
2 CA CA CA CA CA CA
3 OR OR OR OR OR OR
4 NY NY NY NY NY NY
5 SC SC SC TN TN SC
Top Five States in Wind Capacity Additions (2018–2023)
rank 2018 2019 2020 2021 2022 2023
1 TX TX TX TX TX TX
2 IA IA IA IA IA IA
3 OK OK OK OK OK OK
4 CA KS KS KS KS KS
5 KS CA IL IL IL IL

Tables 3–5 summarize the top five states contributing to wind, solar, and hydroelectric capacity additions from 2018 to 2023. Across all three technologies, clear and stable geographic patterns occur. For solar capacity, California consistently ranks first every year, reflecting the state’s long-standing policy incentives and strong solar resource potential (Table 3). North Carolina and Arizona follow closely, though their ranking shifts slightly over time. To be noticed, Texas rises into the top three by 2020, indicating that the state is also having a rapid solar deployment in recent years. Hydroelectric capacity, on the other hand, remains almost entirely dominated by Pacific Northwest and East Coast states (Table 4). Washington is the top hydro state across all six years, with California, Oregon, and New York consistently filling the next three positions. The stability of these rankings proves that hydropower is strongly dependent on established infrastructures and geographic constraints, with relatively few new installations occurring within the study period. Wind capacity growth is concentrated in the central United States. As shown in Table 5, Texas leads wind capacity additions by a wide margin every year, maintaining its position as the largest wind-producing state in the U.S. Iowa and Oklahoma remain strong contributors, holding the second and third positions throughout the period. Kansas and California round out the top five, with Illinois entering the ranking from 2020 onward. These trends support the hypothesis that wind energy development is anchored in states with large available land areas, consistent wind resources, and established transmission networks.

Summary and Conclusions

References

<add references here if relevant, otherwise delete this section>